Enterprise-AI-Search-and-Research-Assistant

Enterprise AI Search & Automation Engine

A Python-based Retrieval-Augmented Generation (RAG) application that:

• ingests PDF and text documents
• chunks and embeds them with OpenAI Embeddings
• stores vectors in a local ChromaDB database
• answers grounded questions with citations
• drafts summary emails through OpenAI function calling
• exposes a Streamlit CIS dashboard with systems logging

Features

• Local document ingestion for .pdf and .txt
• Persistent vector store backed by ChromaDB
• Citation-aware RAG answers
• Structured automation for summary email drafting
• Cost and latency tracking for each OpenAI API call
• Editable summary email drafts with optional direct SMTP sending
• Optional Supabase cloud database and private file storage support

Quickstart

1. Create a virtual environment and install dependencies:

python -m venv venv
venv\Scripts\Activate.ps1
pip install -r requirements.txt

If you already have a venv/ folder, skip creation and just activate it.

1. Copy .env.example to .env and set your API key.

Optional:

• configure SMTP settings in .env if you want users to send summary emails directly from the platform
• configure Supabase keys if you want cloud auth, cloud database, and Supabase Storage
• create a private Supabase Storage bucket named documents or set SUPABASE_STORAGE_BUCKET to your bucket name

1. Run the dashboard:

streamlit run app.py

Project Structure

app.py
src/enterprise_ai_search/
  automation.py
  config.py
  ingestion.py
  logging_utils.py
  models.py
  rag.py
  text_utils.py
  vector_store.py
data/
  chroma/